Bypass capacitor!

Hello! I have a question for the pro here. Right now I am using the NXV200 from Aussie Amplifiers, power supply is 45vdc x2, and 20,000uf (mundorf caps)per rail. The sound is wonderful, no complaints at all I am very happy like it is now. But I have read a lot about Bypassing the big caps with some 22uf Solen caps, would this "extra" improve something in the sound?. Please do not say "get some caps and see for yourself" answer. I just need to know if this "extra" will improve the sound or something else. Thanks for your suggestions and comments.
I have read that in other amps some people say it is a big improvement putting a bypass cap and some say it is just a waist of time and money.
 

bob91343

Member
2010-03-11 10:43 pm
I wouldn't bother. If you had some radio frequencies floating around there, I would consider some small capacitors, nothing as large as 22 microfarads though.

If you want to get serious about this, make a capacitor tester that can measure impedance at the high end of the audio frequency range and see what your filter capacitors do. Or check their data sheet to see impedance curves. It they don't go up high past 10 kHz or so, all is well as it is.
 

djk

R.I.P
2001-02-04 4:23 am
USA
A lot of large computer grade capacitors have 'muddy' sounding bass.

A small bypass capacitor in the range of 22µF~100µF can make the bass sound much 'tighter'.

"Please do not say "get some caps and see for yourself" answer. "

Bad attitude.

"I just need to know if this "extra" will improve the sound or something else."

It will probably change the sound. If you like 'warm' (muddy) sounding bass, you probably won't like adding bypass caps. If you own a Lateral-FET amplifier you may not hear as big of a change as if your amplifier has BJT outputs.
 

Mooly

Administrator
Paid Member
2007-09-15 8:14 am
I can't see them making the slightest difference at all tbh.

That's not to say PSU decoupling isn't important... it is... but it needs to be done correctly at a more local level on the pcb's, not as add ons but as part of the initial design concept.

Why not build a simple high pass filter, say a 0.047uf and 10k to ground directly across one of the main caps and scope the output directly across the 10k then add the extra cap and see if the noise alters. And if it did, which I doubt, then the amp should have it's own on board filtering to stop any audible problems, and should have a high enough PSRR anyway.

:)
 
............... not as add ons but as part of the initial design concept.

........... build a simple high pass filter, say a 0.047uf and 10k to ground directly across one of the main caps and scope the output directly across the 10k then add the extra cap and see if the noise alters. And if it did, which I doubt, then the amp should have it's own on board filtering to stop any audible problems, and should have a high enough PSRR anyway.

:)
you must be capable of measuring the effects of modifying the PSU.
If you can't see the effects, you could be making the PSU and/or amplifier unstable without realising why it sounds different.
An unstable amp sounds quite different from the same amp that is compensated correctly.
 

djk

R.I.P
2001-02-04 4:23 am
USA
""If you own a Lateral-FET amplifier you may not hear as big of a change as if your amplifier has BJT outputs. "

Can you give us some e more detials as to why this is the case? "

My experience with Hitachi Lateral-FETs has been that they sound 'warmer' than BJT. I think this is because they have an Rds on over temperature of about 1R7, and as you hit them hard this increases.

"I will wait if more people have more comments in this matter. "

Bad attitude, again.

If you like 'warm' (muddy) bass, do nothing. If you like 'tight' bass, give them a try.
 
""If you own a Lateral-FET amplifier you may not hear as big of a change as if your amplifier has BJT outputs. "

Can you give us some e more detials as to why this is the case? "

My experience with Hitachi Lateral-FETs has been that they sound 'warmer' than BJT. I think this is because they have an Rds on over temperature of about 1R7, and as you hit them hard this increases.

"I will wait if more people have more comments in this matter. "

Bad attitude, again.

If you like 'warm' (muddy) bass, do nothing. If you like 'tight' bass, give them a try.


Hey djk what is wrong with your "bad attitude" thing ?. I guess you are in your days!.... and That is attitude buddy!!!!:Present:
 
Like somebody said,
if you want to improve filtering
you should not do this i power supply part.
You should put such capacitance close to the circuit.
On the PCB, on the rails that feed transistors.


Then the wires/rails will have some little resistance/inductance
which will interact with the capacitance you have added.
You know there are L-C filters and R-C filters.

There are no such filters as only C (only capacitance).

Bypass caps across big electrolyts is a waste of money, in my opinion.
Or money spent in a bad way.

Regarding people who say they can hear those bypass caps,
I say they probably imagine things.
And this belongs to what we call AUDIO MYTHS.
Audio myths are statements, believes that are unconformed, not validated.
Sometimes they don't even have a logical hypothesis to why they should be true.

Regards :) Lineup
 
Like somebody said,
if you want to improve filtering
you should not do this i power supply part.
You should put such capacitance close to the circuit.
On the PCB, on the rails that feed transistors.


Then the wires/rails will have some little resistance/inductance
which will interact with the capacitance you have added.
You know there are L-C filters and R-C filters.

There are no such filters as only C (only capacitance).

Bypass caps across big electrolyts is a waste of money, in my opinion.
Or money spent in a bad way.

Regarding people who say they can hear those bypass caps,
I say they probably imagine things.
And this belongs to what we call AUDIO MYTHS.
Audio myths are statements, believes that are unconformed, not validated.
Sometimes they don't even have a logical hypothesis to why they should be true.

Regards :) Lineup

Thank you and all the other members for comments. I will not touch that power supply. I just wanted to see if I could add this "extra" and improve something. But I will just leave it as it is, since I am happy with the way it is now. Thanks again guys.
 

djk

R.I.P
2001-02-04 4:23 am
USA
"Bypass caps across big electrolyts is a waste of money, in my opinion."

So, does that mean Adcom engineers are idiots, or deaf?
(the GFA 585 uses two 100µF per channel)

So, does that mean Altec engineers are idiots, or deaf?
(the 9440A uses two 10µF per channel)

So, does that mean Rockford Fosgate engineers are idiots, or deaf?
(the Punch amplifiers use two 10µF per channel)

So, does that mean Prof WM Leach is an idiot, or deaf?
(the Leach amplifier uses two 22µF per channel)

So, does that mean Harman Kardon engineers are idiots, or deaf?
(the Citation 16 uses two 10µF film per channel)

So, does that mean Crest engineers are idiots, or deaf?
(the 9001 uses two 150µF+2.2µF film per channel)

So, does that mean Crown engineers are idiots, or deaf?
(the PSA-2 uses two 4.7µF per channel)

"And this belongs to what we call AUDIO MYTHS."

All food tastes the same(if you have a zinc deficiency), all women kiss the same(if you're a eunuch), VHS and Beta look the same(to Ray Charles and Stevie Wonder), and all amplifiers sound the same(to the deaf old men at Stereo Review).

WHO?

Tell the deaf/dumb/blind kid to go play pinball.
 
"Bypass caps across big electrolyts is a waste of money, in my opinion."

So, does that mean Adcom engineers are idiots, or deaf?
(the GFA 585 uses two 100µF per channel)

So, does that mean Altec engineers are idiots, or deaf?
(the 9440A uses two 10µF per channel)

So, does that mean Rockford Fosgate engineers are idiots, or deaf?
(the Punch amplifiers use two 10µF per channel)

So, does that mean Prof WM Leach is an idiot, or deaf?
(the Leach amplifier uses two 22µF per channel)

So, does that mean Harman Kardon engineers are idiots, or deaf?
(the Citation 16 uses two 10µF film per channel)

So, does that mean Crest engineers are idiots, or deaf?
(the 9001 uses two 150µF+2.2µF film per channel)

So, does that mean Crown engineers are idiots, or deaf?
(the PSA-2 uses two 4.7µF per channel)

[snip].

They are smart. If they would leave those caps out, someone would call them deaf or idiots. They don't want that. ;)

jd
 
Abstract
A filter capacitor network includes an inductor connected in series with an aluminum electrolytic capacitor with the inductor and aluminum electrolytic capacitor bypassed with a capacitor having relatively better high frequency performance characteristics than the aluminum electrolytic capacitor. The inductor decouples the aluminum capacitor from the network at middle and upper frequencies so that the bypass capacitor dominates the characteristics of the network in that range. A signal coupling/DC blocking capacitor network of the same configuration is also disclosed. An amplifier design includes the filter network in the power supply. The filter network is also disclosed in an audio signal path for use in signal coupling/DC blocking.

Claims
We claim:

1. An audio amplifier, comprising:

transformer and rectifier means receiving an AC input power signal and providing a rectified output voltage;

a capacitor network receiving said rectified voltage to provide a DC output signal, said network including an aluminum electrolytic capacitor connected in series with an inductor, said aluminum electrolytic capacitor having a relatively high capacitance and an inherent inductance which degrades its response above a certain frequency, said network further including a bypass capacitor connected to bypass said inductor and said electrolytic capacitor, said bypass capacitor having a relatively small capacitance compared to said electrolytic capacitor and relatively less inherent inductance to provide better high frequency response, said inductor having an inductance which substantially decouples said electrolytic capacitor from said network above a selected frequency so that said bypass capacitor dominates the characteristics of said network above said selected frequency;

amplifier circuit means for receiving an audio input signal and amplifying said signal to provide an amplified audio output signal; and

means for connecting said DC power signal to said amplifier circuit means to provide power thereto.

2. Apparatus according to claim 1 further including a resistor connected in parallel with said inductor, the value of said resistor chosen to minimize resonant effects in said network.

3. An audio amplifier signal path for coupling an audio signal with a DC component from one circuit to another, said signal path including a capacitor network for coupling said audio signal from said one circuit to said circuit while blocking said DC component, said network including an inductor connected in series with an aluminum electrolytic capacitor having a relatively high capacitance and an inherent inductance degrading the response of said aluminum capacitor above a certain frequency, said network further including a bypass capacitor of relatively small capacitance and relatively better high frequency performance compared to said aluminum electrolytic capacitor, said inductor having an inductance which substantially decouples said aluminum electrolytic capacitor from said network above a selected frequency so that said bypass capacitor dominates the characteristics of said network above said selected frequency, said inductor, aluminum electrolytic capacitor, and bypass capacitor having values which provide that all frequencies of said audio signal are passed as free of attenuation and distortion as possible.

4. Apparatus according to claim 3 further including a resistor connected in parallel with said inductor, said resistor having a resistance which minimizes resonance effects in said network.

5. A capacitor network comprising;

an inductor;

an aluminum electrolytic capacitor having a relatively high capacitance and a relatively high inherent inductance, said aluminum electrolytic capacitor connected in series with said inductor;

a bypass capacitor connected to bypass said inductor and aluminum electrolytic capacitor, said bypass capacitor having relatively lower capacitance than said aluminum electrolytic capacitor and having a relatively better high frequency performance; and

said inductor having an inductance which substantially decouples said aluminum electrolytic capacitor from said network above a selected frequency so that said bypass capacitor dominates the characteristics of said network above said frequency.

6. A signal coupling/DC blocking capacitor network, comprising:

an inductor;

an aluminum electrolytic capacitor having a relatively high capacitance and a relatively high inherent inductance, said aluminum electrolytic capacitor connected in series with said inductor;

a bypass capacitor connected by bypass said inductor and aluminum electrolytic capacitor, said bypass capacitor having relatively lower capacitance than said aluminum electrolytic capacitor and having a relatively better high frequency performance;

said inductor having an inductance which substantially decouples said aluminum electrolytic capacitor from said network above a selected frequency so that said bypass capacitor dominates the characteristics of said network above said frequency;

means for connecting one end of said network to an audio signal with a DC component so that said audio signal is coupled through to the other end of said network while said DC component is blocked; and

said inductor, aluminum electrolytic capacitor, and bypass capacitor having values which provide that all frequencies of said audio signal are passed as free of attenuation and distortion as possible.
 
Amplifier bypass capacitors

Different capacitor types, each have their own sonic traits. I have not tried every type for amplifiers, but I have listened to at least 50 different ones for signal (line level).

For amplifier power supply bypass, I have found a large polypropylene, plus a smaller oil filled gives a good sonic improvement.

In one amp, I put a 50uF poly cap, a 5uF poly, and a 1uF oil filled, across each channel. This resulted in quite a sonic improvement. The sound is fuller, more present, and clearer.

In another amp with less room, I put a 20uF metalized poly, and a .47 Jensen oil filled. There wasn't a lot of difference in this amp, but it did improve the sound somewhat.

Don't bypass just for an RF reduction, unless you are building your own amp. The designers of the amps and DIY kits have already done this testing in most cases.

Do not put the Poly caps in digital power supplies. They sound terrible there (from my experience). Use an oil filled capacitor for a sonic improvement. I did this using the same value of capacitors, both were .68uF. The polypropylene capacitor made the sound worse, the oil capacitor made the sound quite better and less harsh.